ABSTRACT. Objective: To study the relationship between 4 components of physical activity and the 12-year incidence of clinical knee osteoarthritis (OA) among older adults. Methods: Longitudinal data from 1,678 men and women, ages 55-85 years, were collected in the Longitudinal Aging Study Amsterdam. Incident clinical knee OA was defined by an algorithm using self-report and general practitioner data. Physical activity was assessed by a validated questionnaire from which 4 physical activity component scores were created: muscle strength, intensity, mechanical strain, and turning actions. Cox proportional hazards models were conducted to examine the relationship between these scores and incident knee OA and reported as hazard ratios (HRs) with 95% confidence intervals (95% CIs). Results: During 12 years of followup, 463 respondents (28%) developed clinical knee OA. A high mechanical strain score (HR 1.43, 95% CI 1.15-1.77) and a low muscle strength score (HR 1.30, 95% CI 1.01-1.68) were associated with an increased risk of knee OA after adjustment for age, sex, region of living, education, lifetime physical work demands, lifetime general physical activity, body mass index, current total physical activity level, and depression. No association was observed in the intensity and turning actions components. The results were similar for men and women, and for obese and nonobese respondents. Conclusion: Older adults who perform low muscle strength activities or activities causing high mechanical strain had an increased risk of clinical knee OA. These results suggest that specific components of physical activity may influence the development of knee OA.